impact of unit characteristics, service configuration and
trust-level characteristics
Introduction
Overall rates of medical intervention during labour and childbirth are increasing in many countries,
although they are levelling off in some where rates are highest.20–22Information about intervention rates in
individual maternity units in the UK is available online to inform women’s decision-making about place of birth,17,23and has also been proposed as a basis for quality indicators, with the recognition that it is
important to understand the potential sources of variation in these rates.18,24
Maternal characteristics and clinical risk factors of women planning birth differ between hospitals and can affect intervention rates, but studies of OU intrapartum intervention rates in England indicate that these factors alone explain only a small part of the wide variation seen.18,25,26These studies have focused on
births occurring in OUs, rather than planned in OUs, and may also include births occurring in associated AMUs as staffing and routine data collection in OUs and AMUs are not always separated and distinct. We are not aware of any previous study exploring variation in intervention rates in births planned in OUs or in births planned in other settings. Analyses presented in this chapter explore variation in unit intervention rates in ‘low risk’ women planning birth in each institutional setting (OU, AMU and FMU) and in NHS trusts providing home birth services and, first, investigate the extent to which this variation can be explained by maternal characteristics.
Studies have also explored the impact of organisational factors, including obstetric and midwifery staffing,27unit size and the level of specialist care available,28–30on intervention rates during labour and
birth. These have also focused on intervention rates in births occurring in OUs, rather than in other settings for birth, and have shown inconclusive or mixed results and that there is little evidence from the UK. In recent years there has been a move towards fewer, larger OUs in England,31with the assumption that the
increased consultant presence that is possible in larger units will improve quality and safety.32Midwifery
staffing levels (midwives per birth) have been increasing since 2008, but in 2012 there were fewer midwives per birth than in 2002.32A recent national survey found that 63% of trusts had fewer midwives
per birth than the recommended minimum level, and only 78% of maternity units reported that they were achieving one-to-one care in labour 90% of the time.32Higher midwifery staffing levels are believed to
improve outcomes and possibly reduce interventions, but the existing evidence on this is limited.33The
extent to which variation in intervention rates can be understood by considering differences in characteristics of the unit or NHS trust, including the number of births, staffing and, for FMUs, the distance to the nearest OU, is also considered in this chapter.
Finally, recent years have also seen changes in the configuration of maternity services. In 2007, in England, two-thirds of NHS trusts providing maternity care did so in OUs only.6The proportion of OUs in England
with an attached AMU has increased, from less than 20% in 2007 to 30% in 2010 and 53% in 2013,6,32
and there has been an increase in the proportion of births in MUs, from 4% in 2006–7 to 11% in 2012.32
Little appears to be known about the possible effects of the configuration of care on intervention rates, either at a trust level or in individual units. It is plausible, for example, that more women opting for non-OU births might change the case mix of women giving birth in OUs, possibly resulting in a ‘higher risk culture’ developing in OUs. Alternatively, the NHS trusts offering more midwifery-led birth options could be
those with a culture of promoting ‘normal birth’ in all maternity settings. We therefore also aimed to explore the extent to which specified aspects of service configuration might explain variation in intervention rates in planned births in OUs.
The results of some of the analyses relating to OUs described and discussed below have also been published elsewhere.34
Methods
Study population
The main study population for these analyses was ‘low risk’ women with a term pregnancy (37 to 42 + 0 weeks’ gestation) planning a vaginal birth in any of the 43 AMUs, 53 FMUs, or 142 NHS trusts providing home birth services, and a stratified random sample of 36 OUs in which data collection for the Birthplace cohort study took place.
Unit or NHS trust characteristics
Using the methods described in Derivation of unit or NHS trust characteristics we derived the variables summarised in Table 5 to describe units or NHS trusts and configuration of care and considered these as factors which might be associated with the study outcomes.
TABLE 5 Unit/NHS trust or configuration characteristics considered as factors that might be associated with outcome measures in different settings
Unit or trust characteristic OU AMU FMU Home
Size (number of births) ✓ ✓ ✓ ✓
Number of delivery beds or bed spaces ✓ ✓ ✓ ✗
Presence of an AMU ✓ ✗ ✗ ✗
% of births in the trust taking place at home ✗ ✗ ✗ ✓
% of births in the trust planned outside the OU ✓ ✗ ✗ ✗
% of births in the trust planned ‘out of hospital’ ✓ ✗ ✗ ✗
Midwifery ‘understaffing’ ✓ ✓ ✓ ✗
Mean number of midwives on duty ✗ ✓ ✓ ✗
Mean number of midwives on duty per woman in labour ✗ ✓ ✓ ✗
Distance to nearest OU ✗ ✗ ✓ ✗
Estimated travel time to nearest OU ✗ ✗ ✓ ✗
Median transfer time to nearest OU ✗ ✗ ✓ ✗
24-hour staffing ✗ ✗ ✓ ✗
Size index ✗ ✓ ✓ ✗
Midwifery staffing index ✗ ✓ ✓ ✗
Distance index ✗ ✗ ✓ ✗
VARIATION BETWEEN UNITS IN INTERVENTIONS AND MATERNAL OUTCOMES IN ‘LOW RISK’ WOMEN
NIHR Journals Library www.journalslibrary.nihr.ac.uk
Derivation of unit or NHS trust characteristics
Size (number of births)
For OUs we used ONS data on births in 2009–10 to derive the number of births per year in each hospital. Where there was an AMU in the same hospital as the OU, these data included births in both settings. Using Birthplace monthly logs of planned births in each unit and cohort study data on transfers before birth, we estimated the annual number of births in each AMU and subtracted this from the total number of births in the hospital to estimate the numbers of births in the OU.
For AMUs and FMUs: we used Birthplace monthly logs of planned births to estimate the annual number of women planning birth in each unit by dividing the number of planned births by the duration of data collection for that unit in years.
For home births: for each NHS trust we used data from the 2007 Healthcare Commission Maternity Services Review on the number of home births in the trust.11
Number of delivery beds or bed spaces
As another indication of the size of a unit, we used data from the 2010 mapping survey on the number of delivery beds or ‘bed spaces’ in the unit on 31 March 2010.6For units that did not reply to the 2010
survey, we used data from the 2007 Healthcare Commission Maternity Services Review.11
Presence of an alongside midwifery unit
An OU was defined as having an AMU if the associated AMU was open for the whole of the period when cohort study data for the OU were being collected.
Percentage of births in each NHS trust taking place at home
For each NHS trust we used data from the 2007 Healthcare Commission Maternity Services Review on the number of home births and the total number of births in the trust to calculate the percentage of births for which intrapartum care was provided by the trust, but took place at home.11
Percentage of births in each NHS trust planned outside an obstetric unit and ‘out of hospital’
We used Birthplace monthly logs of planned births in units and at home to calculate the number of births planned to take place outside an OU (i.e. in an AMU, in a FMU or at home) and ‘out of hospital’ (i.e. in a FMU or at home) in each NHS trust. The total number of births in the NHS trust was calculated by summing ONS data on maternities in 2009–10 for each of the OUs and MUs in the trust (OU, AMU and FMU) and adding to this the estimated annual number of home births in the trust (planned minus transferred) from Birthplace data. Unplanned home births were excluded from both the numerator and the denominator.
Midwifery staffing levels and ‘understaffing’
For OUs we used data from the Birthplace staffing logs to estimate the proportion of shifts with at least one woman in the unit where the total number of women in the delivery suite or labour ward exceeded the number of midwives on duty as a measure of midwifery ‘understaffing’.
For AMUs and FMUs: we calculated the mean number of midwives on duty and the mean number of midwives on duty per woman in labour for shifts when there was at least one woman in labour for each of the AMUs and FMUs. We also estimated midwifery ‘understaffing’ in AMUs and FMUs using the proportion of shifts with at least one woman in labour where the number of women in labour exceeded the number of midwives on duty.
The Birthplace staffing log sheets from each unit included data recorded for two shifts, with data points at 09.00 and 21.00. Proportions of shifts ‘understaffed’ included all morning and evening shifts at each unit for which data were available.
Distance to nearest obstetric unit
For each FMU we used postcode data and Google Maps to calculate the distance and estimated travel time by road to the nearest OU in the same trust. We used Birthplace data on transfer times to calculate the median time from the start of transfer to the start of care in an OU for women transferred from each FMU.
Twenty-four-hour staffing
We used data from the 2010 Birthplace mapping survey6(see Appendix 4) or, if no data were available,
the 2007 Healthcare Commission Maternity Services Review11to identify whether or not each FMU was
staffed 24 hours per day.
Size, staffing and distance indices
Because several variables in our data set may have been interpreted as indicators of the same underlying characteristics of the units we combined these into single indices for size, staffing and, for FMUs only, distance from the nearest OU, using methods described in Statistical methods, Size, staffing and distance
indices. The size index used data on the number of delivery beds, the number of women planning birth in
the unit per year and the mean number of midwives on duty when there was at least one woman in labour. The staffing index used the mean number of midwives on duty per woman in labour and the percentage of shifts ‘understaffed’. The distance index (for FMUs) was based on distance and travel time to the nearest OU measured using Google Maps and the median transfer time for that unit from the
Birthplace cohort study. Larger indices indicate larger units, higher staffing levels and units further away from the nearest OU.
Outcome measures
We used the following four main ‘outcome measures’, two capturing interventions during labour and birth, one indicating birth without complications that might affect future births and one indicating birth with low intervention:
l intrapartum caesarean section
l instrumental delivery (forceps or ventouse)
l ‘straightforward birth’, defined as birth without forceps or ventouse, intrapartum caesarean section,
third- or fourth-degree perineal trauma or blood transfusion
l ‘normal birth’, defined as birth without induction of labour, epidural or spinal analgesia, general
anaesthetic, forceps or ventouse, caesarean section or episiotomy.12
For each we calculated adjusted unit or trust (for home birth) rates. For analyses of OU intervention rates we also used the following additional ‘outcome measures’, not included in the original planned analyses, to help to explain observations from the analyses of our main outcome measures:
l epidural or spinal analgesia
l augmentation with oxytocin.
Unit/trust rates of each intervention or outcome were adjusted for the maternal characteristics listed in
Chapter 2 (see Maternal characteristics) and for the presence of one or more ‘complicating conditions’
identified at the start of care in labour, as described in Chapter 2 (see ‘Complicating conditions’ at the start
of labour care).
VARIATION BETWEEN UNITS IN INTERVENTIONS AND MATERNAL OUTCOMES IN ‘LOW RISK’ WOMEN
NIHR Journals Library www.journalslibrary.nihr.ac.uk
Statistical methods
Calculation of intervention and outcome rates
We calculated unadjusted (observed) event rates in each unit or NHS trust (for home births) separately for nulliparous and multiparous women by dividing the number of women with a given intervention or outcome by the total number of women in the Birthplace data set in that unit. Adjusted unit-level event rates were calculated using an indirect standardisation procedure (see Chapter 2, Calculation of adjusted
rates for individual units or trusts).18,19
Variation between units/trusts in rates of maternal interventions or outcomes
We plotted the adjusted event rates against the numbers of women in the Birthplace sample on funnel plots with 95% and 99.8% control limits as described in Chapter 2 (see Funnel plots).19The control limits,
which represent approximately two and three standard deviations, respectively, around the overall mean event rate, were used to assess whether or not there was more variation than might be expected by chance after allowing for differences in maternal characteristics between units.19
Size, staffing and distance indices
Indices of size, staffing and distance from the nearest OU were created by combining several variables indicative of each of these characteristics as described in Derivation of unit or NHS trust characteristics,
Size, staffing and distance indices. Principal components analysis was used to combine the measurements
in each index, with the score on the first principal component being taken as the measurement for that unit’s size, staffing or distance from an OU. These ‘measurements’ have no units, but indicate which units are bigger, better staffed or further from an OU than other units in the data set.
Effects of unit or configuration characteristics on intervention and outcome rates
Simple linear regression was used to investigate whether unit or configuration characteristics were associated with variations in the study outcomes (the adjusted event rates). The adjusted event rates were regressed on each of the unit characteristics in turn. Robust standard errors were used to take account of non-constant variance among the outcome rates with increases in some of the unit characteristics (heteroscedasticity). The regressions were also weighted to take account of the number of observations used to calculate each unit’s event rate.
Further exploratory analyses
We carried out a series of post-hoc analyses to further explore some of the associations found in the analyses relating to OUs. First, because of their possible association with other interventions we carried out additional analyses of rates of augmentation and epidural or spinal analgesia use. Second, we explored whether or not the proportion of planned births in an OU that were ‘higher risk’ (estimated from the Birthplace cohort) had an impact on intervention rates in planned ‘low risk’ births in OUs. Third, we investigated whether or not intervention rates were correlated in ‘low risk’ and ‘higher risk’ women within the same OU. Pearson’s correlation coefficients were used to describe the strength of association between rates and p-values were calculated after verifying approximate Gaussian distributions of the variables. Finally, we examined whether or not OUs situated in trusts with a higher percentage of non-OU births tended to be those with an associated AMU.
Sensitivity analyses
We plotted significant relationships between unit or configuration characteristics and outcome measures on scatter graphs; where these suggested that the relationships might have been heavily dependent on a small number of outliers, sensitivity analyses were carried out by repeating the regressions while excluding the outliers identified on the graphs.
Results
Characteristics of obstetric units in the study sample
For 1 of the 36 OUs included in the study we had insufficient data on the associated AMU to enable us to estimate the number of births in the OU. The remaining OUs varied in size, as measured by number of births per year, and in the number of delivery beds (Table 6). Data on midwifery staffing were available for only 30 OUs, but a relatively high proportion (median 30%) of shifts in those OUs were ‘understaffed’, with less than one midwife for each woman in labour. For six OUs it was not possible to calculate the proportion of non-OU and ‘out-of-hospital’ births because insufficient data were available to estimate the annual number of planned FMU births (four trusts) or home births (two trusts). In line with national figures, a relatively small proportion of births were planned to take place outside OUs and ‘out of hospital’, but with some variation between trusts. Nine OUs had an associated AMU in the same hospital.
TABLE 6 Characteristics of the OUs in the Birthplace study
Unit or configuration characteristic n Median (IQR) Min. Max.
Sizea 35
2919 (2361–3849) 1380 6490
Number of delivery bedsb
36 10 (8–12) 5 19
% midwifery ‘understaffing’c 30
29.6 (20.5–41.8) 4.4 83.6
% of planned non-OU birthsd
30 3.0 (2.3–7.9) 0.4 37.2
% of planned ‘out-of-hospital’ birthse 30
2.4 (1.4–4.1) 0.4 10.2
IQR, interquartile range; max., maximum; min., minimum.
a Number of births in the OU (excluding those taking place in any associated AMU) April 2009 to March 2010. b Number of delivery beds or bed spaces in the OU.
c Percentage of shifts where there was less than one midwife on duty per woman in the delivery suite. d Percentage of births in the NHS trust planned to take place at home, in a FMU or in an AMU. e Percentage of births in the NHS trust planned to take place at home or in a FMU.
VARIATION BETWEEN UNITS IN INTERVENTIONS AND MATERNAL OUTCOMES IN ‘LOW RISK’ WOMEN
NIHR Journals Library www.journalslibrary.nihr.ac.uk
Variation in intervention rates between obstetric units
We examined intervention rates, both unadjusted and adjusted for differences in maternal characteristics between OUs, to study the variation between OUs. Adjustment generally had little effect on the rates (see example in Figure 2) and so subsequent figures and tables all use adjusted rates.
For all study outcomes, funnel plots showed that there was more variation in adjusted intervention rates in nulliparous and multiparous women planning birth in OUs than would be expected by chance alone (Figures 3 and 4). There was more variation in intervention rates in nulliparous women than in multiparous women for all outcomes except ‘normal birth’, where the variation in rates in nulliparous and multiparous women was similar (Table 7). There was more variation in rates of ‘normal birth’ than for our other maternal outcome, ‘straightforward birth’. For most outcomes the number of OUs with intervention rates that were higher or lower than would be expected by chance was broadly similar (see Figures 3 and 4).
Configuration, unit characteristics and intervention rates in obstetric units
There was no significant association between the number of OU delivery beds or the percentage of births in the trust that were planned ‘out of hospital’ and any of the main outcome measures studied (Table 8).
There was a significant association between the OU size (number of births) and the intrapartum caesarean section rate in planned OU births; larger OUs had lower intrapartum caesarean section rates in both nulliparous and multiparous women. Larger OUs also had a significantly higher ‘straightforward birth’ rate